@inproceedings{chi-etal-2025-modeling,
title = "{M}ode{L}ing: A Novel Dataset for Testing Linguistic Reasoning in Language Models",
author = "Chi, Nathan Andrew and
Malchev, Teodor and
Kong, Riley and
Chi, Ryan Andrew and
Huang, Lucas and
Chi, Ethan A and
McCoy, R. Thomas and
Radev, Dragomir",
editor = "Ojha, Atul Kr. and
Liu, Chao-hong and
Vylomova, Ekaterina and
Pirinen, Flammie and
Washington, Jonathan and
Oco, Nathaniel and
Zhao, Xiaobing",
booktitle = "Proceedings of the Eighth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2025)",
month = may,
year = "2025",
address = "Albuquerque, New Mexico, U.S.A.",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.loresmt-1.10/",
doi = "10.18653/v1/2025.loresmt-1.10",
pages = "105--114",
ISBN = "979-8-89176-230-5",
abstract = "We introduce ModeLing, a novel benchmark of Linguistics Olympiad-style puzzles which tests few-shot reasoning in AI systems. Solving these puzzles necessitates inferring aspects of a language{'}s grammatical structure from a small number of examples. Such puzzles provide a natural testbed for language models, as they require compositional generalization and few-shot inductive reasoning. Consisting solely of new puzzles written specifically for this work, ModeLing has no risk of appearing in the training data of existing AI systems: this ameliorates the risk of data leakage, a potential confounder for many prior evaluations of reasoning. Evaluating several large open source language models and GPT on our benchmark, we observe non-negligible accuracy, demonstrating few-shot emergent reasoning ability which cannot merely be attributed to shallow memorization. However, imperfect model performance suggests that ModeLing can be used to measure further progress in linguistic reasoning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chi-etal-2025-modeling">
<titleInfo>
<title>ModeLing: A Novel Dataset for Testing Linguistic Reasoning in Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nathan</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Chi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Teodor</namePart>
<namePart type="family">Malchev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Riley</namePart>
<namePart type="family">Kong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="given">Andrew</namePart>
<namePart type="family">Chi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucas</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ethan</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Chi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">R</namePart>
<namePart type="given">Thomas</namePart>
<namePart type="family">McCoy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dragomir</namePart>
<namePart type="family">Radev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Eighth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Atul</namePart>
<namePart type="given">Kr.</namePart>
<namePart type="family">Ojha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chao-hong</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Vylomova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Flammie</namePart>
<namePart type="family">Pirinen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Washington</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nathaniel</namePart>
<namePart type="family">Oco</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaobing</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico, U.S.A.</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-230-5</identifier>
</relatedItem>
<abstract>We introduce ModeLing, a novel benchmark of Linguistics Olympiad-style puzzles which tests few-shot reasoning in AI systems. Solving these puzzles necessitates inferring aspects of a language’s grammatical structure from a small number of examples. Such puzzles provide a natural testbed for language models, as they require compositional generalization and few-shot inductive reasoning. Consisting solely of new puzzles written specifically for this work, ModeLing has no risk of appearing in the training data of existing AI systems: this ameliorates the risk of data leakage, a potential confounder for many prior evaluations of reasoning. Evaluating several large open source language models and GPT on our benchmark, we observe non-negligible accuracy, demonstrating few-shot emergent reasoning ability which cannot merely be attributed to shallow memorization. However, imperfect model performance suggests that ModeLing can be used to measure further progress in linguistic reasoning.</abstract>
<identifier type="citekey">chi-etal-2025-modeling</identifier>
<identifier type="doi">10.18653/v1/2025.loresmt-1.10</identifier>
<location>
<url>https://aclanthology.org/2025.loresmt-1.10/</url>
</location>
<part>
<date>2025-05</date>
<extent unit="page">
<start>105</start>
<end>114</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ModeLing: A Novel Dataset for Testing Linguistic Reasoning in Language Models
%A Chi, Nathan Andrew
%A Malchev, Teodor
%A Kong, Riley
%A Chi, Ryan Andrew
%A Huang, Lucas
%A Chi, Ethan A.
%A McCoy, R. Thomas
%A Radev, Dragomir
%Y Ojha, Atul Kr.
%Y Liu, Chao-hong
%Y Vylomova, Ekaterina
%Y Pirinen, Flammie
%Y Washington, Jonathan
%Y Oco, Nathaniel
%Y Zhao, Xiaobing
%S Proceedings of the Eighth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2025)
%D 2025
%8 May
%I Association for Computational Linguistics
%C Albuquerque, New Mexico, U.S.A.
%@ 979-8-89176-230-5
%F chi-etal-2025-modeling
%X We introduce ModeLing, a novel benchmark of Linguistics Olympiad-style puzzles which tests few-shot reasoning in AI systems. Solving these puzzles necessitates inferring aspects of a language’s grammatical structure from a small number of examples. Such puzzles provide a natural testbed for language models, as they require compositional generalization and few-shot inductive reasoning. Consisting solely of new puzzles written specifically for this work, ModeLing has no risk of appearing in the training data of existing AI systems: this ameliorates the risk of data leakage, a potential confounder for many prior evaluations of reasoning. Evaluating several large open source language models and GPT on our benchmark, we observe non-negligible accuracy, demonstrating few-shot emergent reasoning ability which cannot merely be attributed to shallow memorization. However, imperfect model performance suggests that ModeLing can be used to measure further progress in linguistic reasoning.
%R 10.18653/v1/2025.loresmt-1.10
%U https://aclanthology.org/2025.loresmt-1.10/
%U https://doi.org/10.18653/v1/2025.loresmt-1.10
%P 105-114
Markdown (Informal)
[ModeLing: A Novel Dataset for Testing Linguistic Reasoning in Language Models](https://aclanthology.org/2025.loresmt-1.10/) (Chi et al., LoResMT 2025)
ACL
- Nathan Andrew Chi, Teodor Malchev, Riley Kong, Ryan Andrew Chi, Lucas Huang, Ethan A Chi, R. Thomas McCoy, and Dragomir Radev. 2025. ModeLing: A Novel Dataset for Testing Linguistic Reasoning in Language Models. In Proceedings of the Eighth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2025), pages 105–114, Albuquerque, New Mexico, U.S.A.. Association for Computational Linguistics.